U.S. patent number 9,484,771 [Application Number 14/152,392] was granted by the patent office on 2016-11-01 for uninterruptable power supply for device having power supply modules with internal automatic transfer switches.
This patent grant is currently assigned to Juniper Networks, Inc.. The grantee listed for this patent is Juniper Networks, Inc.. Invention is credited to Michael Braylovskiy, Jaspal Gill, Muhammad Sagarwala.
United States Patent |
9,484,771 |
Braylovskiy , et
al. |
November 1, 2016 |
Uninterruptable power supply for device having power supply modules
with internal automatic transfer switches
Abstract
Techniques are described for determining whether power from a
first power source is unavailable to a power supply module. In
response to determining that power from the first power source is
unavailable, the techniques de-couple the first power source from
one or more components of an electronic device connected to an
output of the power supply module with one or more de-coupling
components of the power supply module that connect an automatic
transfer switch (ATS) of the power supply module to an output of
the power supply module. Subsequent to de-coupling the first power
source from the one or more components of the electronic device,
the techniques de-couple a power supply module from the first power
source. The techniques couple the power supply module to a second
power source for delivering power to the one or more components of
the electronic device.
Inventors: |
Braylovskiy; Michael (San
Mateo, CA), Gill; Jaspal (Tracy, CA), Sagarwala;
Muhammad (Los Gatos, CA) |
Applicant: |
Name |
City |
State |
Country |
Type |
Juniper Networks, Inc. |
Sunnyvale |
CA |
US |
|
|
Assignee: |
Juniper Networks, Inc.
(Sunnyvale, CA)
|
Family
ID: |
53006510 |
Appl.
No.: |
14/152,392 |
Filed: |
January 10, 2014 |
Prior Publication Data
|
|
|
|
Document
Identifier |
Publication Date |
|
US 20150123473 A1 |
May 7, 2015 |
|
Related U.S. Patent Documents
|
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
Issue Date |
|
|
61898974 |
Nov 1, 2013 |
|
|
|
|
Current U.S.
Class: |
1/1 |
Current CPC
Class: |
H02J
9/061 (20130101); H02J 1/102 (20130101) |
Current International
Class: |
H02J
9/06 (20060101); H02J 1/10 (20060101); H02J
3/00 (20060101) |
Field of
Search: |
;307/64 |
References Cited
[Referenced By]
U.S. Patent Documents
Other References
"Information technology equipment--Safety--Part 1: General
requirements," International Standard IEC 60950-1, Second Edition,
Dec. 2005, 312 pp. cited by applicant .
U.S. Appl. No. 13/946,916, by David K. Owen, filed Jul. 19, 2013.
35 pp. cited by applicant.
|
Primary Examiner: Deberadinis; Robert
Attorney, Agent or Firm: Shumaker & Sieffert, P.A.
Parent Case Text
This application claims the benefit of U.S. Provisional Application
No. 61/898,974, filed Nov. 1, 2013, the content of which is
incorporated by reference in its entirety.
Claims
What is claimed is:
1. A method comprising: determining, with a controller of a power
supply module of an electronic device, whether power from a first
power source is unavailable to the power supply module; responsive
to determining that power from the first power source is
unavailable, de-coupling, with one or more de-coupling components
of the power supply module that connect an automatic transfer
switch (ATS) of the power supply module to an output of the power
supply module, the first power source from one or more components
of the electronic device that are connected to the output of the
power supply module; subsequent to de-coupling the first power
source from the one or more components of the electronic device,
de-coupling, with the ATS, the power supply module from the first
power source; and coupling, with the ATS, the power supply module
to a second power source for delivering power to the one or more
components of the electronic device.
2. The method of claim 1, wherein determining whether power from
the first power source is unavailable comprises: receiving a signal
indicative of at least one of an AC voltage and an AC current at a
connection of the first power source to the power supply module;
and determining that the power from the first power source is
unavailable when one or both of the AC voltage and the AC current
is less than a threshold.
3. The method of claim 1, wherein the one or more de-coupling
components comprise one or more silicone controlled rectifiers
(SCRs) of the power supply module, and wherein de-coupling the
first power source from the one or more components of the
electronic device comprises de-asserting one or more of the
SCRs.
4. The method of claim 1, wherein the one or more de-coupling
components comprise one or more silicone controlled rectifiers
(SCRs) of a bridge rectifier of the power supply module, wherein
de-coupling the first power source from the one or more components
of the electronic device comprises de-asserting one or more of the
SCRs of the bridge rectifier, and wherein the bridge rectifier is
configured to convert negative portions of an AC power of the first
power source to positive portions.
5. The method of claim 1, wherein de-coupling the power supply
module from the first power source comprises toggling one or more
relays the ATS of the power supply module from a close state to an
open state.
6. The method of claim 1, wherein de-coupling the power supply
module from the first power source comprises waiting until current
through the one or more de-coupling components of the power supply
module is approximately zero before de-coupling the power supply
from the first power source.
7. The method of claim 1, further comprising: determining a ride
through period, wherein coupling the power supply module to the
second power source comprises coupling the power supply module to
the second power source based on the determined ride through
period.
8. The method of claim 1, further comprising: determining whether
power from the first power source is available during a ride
through period, wherein de-coupling the first power source from the
one or more components of the electronic device comprises
de-coupling the first power source from the one or more components
of the electronic device if power from the first power source is
unavailable after the ride through period.
9. The method of claim 1, further comprising: after coupling the
power supply module to the second power source, determining whether
power from the first power source is available; responsive to
determining that power from the first power source is available,
de-coupling, with the one or more de-coupling components of the
power supply module, the second power source from the one or more
components of the electronic device that are connected to the
output of the power supply module; subsequent to de-coupling the
second power source from the one or more components of the
electronic device, de-coupling, with the ATS, the power supply
module from the second power source; and coupling, with the ATS,
the power supply module to the first power source for delivering
power to the one or more components of the electronic device.
10. The method of claim 9, wherein de-coupling the power supply
module from the second power source comprises de-coupling, after a
restore period, the power supply module from the second power
source.
11. The method of claim 1, further comprising: recording transfer
events of the ATS; and communicating information indicative of the
transfer events to the electronic device.
12. An electronic device comprising: one or more components; and a
power supply module, the power supply module comprising: an output
connected to the one or more components; an automatic transfer
switch (ATS); one or more de-coupling components that connect the
ATS to the output of the power supply module; and a controller
configured to determine whether power from a first power source is
unavailable to the power supply module, cause the one or more
de-coupling components to de-couple the first power source from the
one or more components of the electronic device that are connected
to the output of the power supply module responsive to determining
that power from the first power source is unavailable, cause the
ATS to de-couple the power supply module from the first power
source subsequent to causing the one or more de-coupling components
to de-couple the first power source from the one or more components
of the electronic device, and cause the ATS to couple the power
supply module to a second power source for delivering power to the
one or more components of the electronic device.
13. The device of claim 12, further comprising: at least one of a
voltage sensor and a current sensor configured to measure an AC
voltage or an AC current, respectively, at a connection of the
first power source to the power supply module, wherein the
controller is configured to receive a signal indicative of the at
least one of the AC voltage or the AC current, and determine that
the power from the first power source is unavailable when one or
both of the AC voltage and the AC current is less than a
threshold.
14. The device of claim 12, wherein the power supply module further
comprises: a silicone controlled rectifier (SCR) control unit,
wherein the one or more decoupling components comprise one or more
SCRs, and wherein the controller is configured to cause the SCR
control unit to de-assert the one or more SCRs to de-couple the
first power source from the one or more components of the
electronic device.
15. The device of claim 12, wherein the power supply module further
comprises: a bridge rectifier configured to convert negative
portions of an AC power of the first power source to positive
portions, wherein the bridge rectifier includes the one or more
de-coupling components, and wherein the one or more de-coupling
components comprise silicone controlled rectifiers (SCRs); and a
SCR control unit, wherein the controller is configured to cause the
SCR control unit to de-assert the one or more SCRs to de-couple the
first power source from the one or more components of the
electronic device.
16. The device of claim 12, further comprising: a relay driver,
wherein the ATS comprises one or more relays, and wherein the
controller is configured to cause the relay driver to toggle the
one or more relays of the ATS from a close state to an open state
to cause the ATS to de-couple the power supply module from the
first power source.
17. The device of claim 12, wherein the controller is configured to
wait until current through the one or more de-coupling units is
approximately zero before causing the ATS to de-couple the power
supply module from the first power source.
18. The device of claim 12, wherein the controller is configured to
determine a ride through period, and wherein to cause the ATS to
couple the power supply module to the second power source, the
controller is configured to cause the ATS to couple the power
supply module to the second power source based on the determined
ride through period.
19. The device of claim 12, wherein the controller is configured to
determine whether power from the first power source is available
during a ride through period, and wherein the controller is
configured to cause the one or more de-coupling components to
de-couple the first power source from the one or more components of
the electronic device if power from the first power source is
unavailable after the ride through period.
20. A power supply module for delivering power to an electronic
device, the power supply module comprising: an output connected to
one or more components of the electronic device; an automatic
transfer switch (ATS); one or more de-coupling components that
connect the ATS to the output of the power supply module; and a
controller configured to determine whether power from a first power
source is unavailable to the power supply module, cause the one or
more de-coupling components to de-couple the first power source
from the one or more components of the electronic device that are
connected to the output of the power supply module responsive to
determining that power from the first power source is unavailable,
cause the ATS to de-couple the power supply module from the first
power source subsequent to causing the one or more de-coupling
components to de-couple the first power source from the one or more
components of the electronic device, and cause the ATS to couple
the power supply module to a second power source for delivering
power to the one or more components of the electronic device.
Description
TECHNICAL FIELD
This disclosure relates to power feeds of a power supply, and more
particularly, to relays coupled to the power feeds and the power
supply.
BACKGROUND
Exponential growth of computer networks around the world, as well
as ever-increasing reliance on those networks, imposes challenging
requirements on reliability of telecommunication equipment
(switches, routers etc.). As such, maintaining uninterruptable
operation of telecommunication equipment is an important
operational requirement for manufacturers of such equipment. One
technique of achieving uninterruptable operation is redundant
configuration of main components of telecommunication equipment,
including utilizing redundant power supply modules (PSMs) and AC
power sources.
Quality of energy provide by an AC power source is often not
perfect. For example, interruption of AC energy flow frequently
occurs due a variety of factors, such as weather condition,
overload of AC lines at peak times, etc. In the event an AC line
interruption lasts longer then a hold-up time of an AC/DC power
supply module (PSM) that supplies power to the telecommunication
equipment, the equipment ceases operation. Many complex
telecommunication devices require significant time to restart once
power is restored. As a result, even a short power disruption may
result in severe interruption of network traffic.
To provide redundant power feeds to an electronic device, AC
distribution may include automatic transfer switches (ATSs). An ATS
couples to a plurality of power feeds to one or more power supplies
of the electronic device. The power supplies receive power from one
of the power feeds at a time via the ATS, convert the received AC
power to DC power to power up the electronic device. If power from
a primary power feed becomes unavailable, the ATS quickly switches
from the primary power feed to a backup power feed so that input
power to the one or more power supplies is uninterrupted. This
switching allows the electronic device to remain operational even
when the primary power feed is unavailable.
SUMMARY
This disclosure describes one or more de-coupling components of a
power supply module of an electronic device that connect an
automatic transfer switch (ATS) of the power supply module to an
output of the power supply module. The de-coupling components
de-couple one or more components of the electronic device connected
to the output of the power supply module from a primary power
source, in response to power from the primary power source being
unavailable, prior to the ATS de-coupling the power supply module
from the primary power source and prior to the ATS coupling the
power supply module to a secondary power source for delivering
power to the one or more components of the electronic device.
In this manner, the techniques described in this disclosure may
ensure that input current into the power supply module is
approximately zero amperes (amps or A) before transitioning the
power supply module from receiving power from the secondary power
source. For example, the techniques may assert or de-assert control
signals to de-coupling components such as rectifiers to control the
input current and ensure that the ATS transitions from the primary
power source to the secondary power source after the input current
is approximately zero amps (i.e., interrupt the input current once
it crosses zero). Although the ATS may transition from one power
source to another source after the input current is approximately
zero amps, the ATS may still transition from one power source to
another within a required holdup time (e.g., without system
interruption).
The techniques described herein may provide certain advantages. For
example, with the power supply module including the ATS, the
techniques mitigate against the chance of a single point failure.
Also, by transitioning from one power source to another after the
input current is approximately zero amps, the techniques may reduce
arcing, overvoltage spikes, and increase life time and reliability
of mechanical relays within the ATS.
In one example, the disclosure describes a method comprising
determining, with a controller of a power supply module of an
electronic device, whether power from a first power source is
unavailable to the power supply module. The method includes
responsive to determining that power from the first power source is
unavailable, de-coupling, with one or more de-coupling components
of the power supply module that connect an automatic transfer
switch (ATS) of the power supply module to an output of the power
supply module, the first power source from one or more components
of the electronic device that are connected to the output of the
power supply module. The method also includes subsequent to
de-coupling the first power source from the one or more components
of the electronic device, de-coupling, with the ATS, the power
supply module from the first power source. The method further
includes coupling, with the ATS, the power supply module to a
second power source for delivering power to the one or more
components of the electronic device.
In one example, the disclosure describes an electronic device
comprising one or more components and a power supply module. The
power supply module includes an output connected to the one or more
components, an automatic transfer switch (ATS), one or more
de-coupling components that connect the ATS to the output of the
power supply module, and a controller. The controller is configured
to determine whether power from a first power source is unavailable
to the power supply module, cause the one or more de-coupling
components to de-couple the first power source from the one or more
components of the electronic device that are connected to the
output of the power supply module responsive to determining that
power from the first power source is unavailable, cause the ATS to
de-couple the power supply module from the first power source
subsequent to causing the one or more de-coupling components to
de-couple the first power source from the one or more components of
the electronic device, and cause the ATS to couple the power supply
module to a second power source for delivering power to the one or
more components of the electronic device.
In one example, the disclosure describes a power supply module for
delivering power to an electronic device. The power supply module
includes an output connected to one or more components of the
electronic device, an automatic transfer switch (ATS), one or more
de-coupling components that connect the ATS to the output of the
power supply module, and a controller. The controller is configured
to determine whether power from a first power source is unavailable
to the power supply module, cause the one or more de-coupling
components to de-couple the first power source from the one or more
components of the electronic device that are connected to the
output of the power supply module responsive to determining that
power from the first power source is unavailable, cause the ATS to
de-couple the power supply module from the first power source
subsequent to causing the one or more de-coupling components to
de-couple the first power source from the one or more components of
the electronic device, and cause the ATS to couple the power supply
module to a second power source for delivering power to the one or
more components of the electronic device.
In one example, the disclosure describes a system comprising means
for determining, with a controller of a power supply module of an
electronic device, whether power from a first power source is
unavailable to the power supply module. The system also includes
means for de-coupling the first power source from one or more
components of the electronic device that are connected to the
output of the power supply module responsive to determining that
power from the first power source is unavailable, wherein the means
for de-coupling connects an automatic transfer switch (ATS) of the
power supply module to an output of the power supply module. The
system further includes means for de-coupling, with the ATS, the
power supply module from the first power source subsequent to
de-coupling the first power source from the one or more components
of the electronic device. The system also includes means for
coupling, with the ATS, the power supply module to a second power
source for delivering power to the one or more components of the
electronic device.
The details of one or more techniques of the disclosure are set
forth in the accompanying drawings and the description below. Other
features, objects, and advantages of the techniques will be
apparent from the description and drawings, and from the
claims.
BRIEF DESCRIPTION OF DRAWINGS
FIG. 1 is a block diagram illustrating an example telecommunication
system.
FIG. 2 is a block diagram illustrating another example
telecommunication system.
FIG. 3 is a block diagram illustrating an example electronic device
configured to implement one or more of the techniques described in
this disclosure.
FIG. 4 is a block diagram illustrating one example of a power
supply module that includes an automatic transfer switch (ATS) in
accordance with one or more examples described in this
disclosure.
FIG. 5 is a timing diagram illustrating techniques in accordance
with one or more examples described in this disclosure.
FIG. 6 is a flowchart illustrating the corresponding timing in FIG.
5.
FIG. 7A is a block diagram illustrating an example of a relay
driver of a power supply module.
FIG. 7B is a block diagram illustrating another example of a relay
driver of a power supply module.
FIG. 8 is a flowchart illustrating an example in accordance with
one or more techniques of this disclosure.
DETAILED DESCRIPTION
FIG. 1 is a block diagram illustrating a telecommunication system
2A that includes alternating current (AC) power sources (1, 2)
coupled to electronic device 10A. Examples of electronic device 10A
include a router, switch, gateway, intrusion protection device,
firewall and the like. For example, electronic device 10A may
route, forward or otherwise transport data in a network, such as a
service provider network or a local area network (LAN). The
techniques described in this disclosure are not limited to examples
where the electronic device is a router, and can be extended to
other types of electronic devices that utilize power feeds for
supplying power.
In telecommunication system 2A, electronic device 10A includes N+N
power supply module (PSM) redundancy scheme with 1+1 redundant AC
power sources. For example, as illustrated, electronic device 10A
includes a first set of N power supply modules coupled to a first
AC power source (i.e., AC Source 1), and a second set of N power
supply modules coupled to a second AC power source (i.e., AC Source
2).
The power supply modules illustrated in FIG. 1 receive AC power
from respective AC sources 1 or 2 and convert the AC power to a DC
power. AC sources 1 and 2 provide alternating-current (AC) power.
As one example, AC sources 1 and 2 provide approximately 200 volts
AC (VAC) to 277 VAC. In some examples, AC source 1 may be a primary
power feed, and AC source 2 may be a redundant, secondary power
feed.
The power supply modules output the DC power to a common bus, for
instance, the Common Vout Bus illustrated in FIG. 1. The components
of electronic device 10A couple to the common bus and receive DC
power from the common bus for operation. For instance, FIG. 1
illustrates one or more cards 12 coupled to the common bus. The one
or more cards 12 receive power from the N+N power supply modules.
Examples of one or more cards 12 include control cards, fabric
cards, and line cards, although other components of electronic
device 10A may also receive power via the common bus.
In this way, if there is loss of power from one of the AC sources,
one or more cards 12 of electronic device 10A may still receive
power. For instance, if AC Source 1 was experiencing an
interruption (e.g., loss of power or brownout), one or more cards
12 may receive power from the second set of power supply modules
coupled to AC Source 2. Similarly, if AC Source 2 was experiencing
an interruption (e.g., loss of power or brownout), one or more
cards 12 may receive power from the second set of power supply
modules coupled to AC Source 1.
However, there may be drawbacks to the system illustrated in FIG.
1. For instance, for redundancy, a total of 2N power supply modules
are needed, where N power supplies deliver power and N power
supplies provide backup power (e.g., N power supplies were coupled
to the first power feed and the other N power supplies were coupled
to the second power feed). Utilizing an N+N power supplies
redundancy scheme results in the power supplies requiring
approximately twice as much space in electronic device 10A and
twice the power supply cost as compared to utilizing N power
supplies or N+1 power supplies, as may be possible using the
techniques described in this disclosure.
Moreover, the N+N PSM redundancy scheme provides poor PSM
utilization. For example, each set of N power supply modules may be
configured to deliver all the power needed to power the components
of electronic device 10A. As one example, if electronic device 10A
requires 1000 Watts (W) of power, and assume that N equals 10, then
each of the N power supply modules in the first set of N power
supply modules is designed to deliver approximately 100 W of power.
Because the N power supply modules of the second set of N power
supply modules supply power to the components of electronic device
10A in case the first AC power source becomes unavailable, the N
power supply modules of the second set of N power supply modules
are also designed to each deliver approximately 100 W of power for
a total of 1000 W of power.
However, if both the first AC power source and the second AC power
source are available, then each power supply module of the N+N
power supply modules deliver 50 W (i.e., 50 W*20 power supply
modules equals 1000 W). In this case, because the maximum power the
can be delivered by a single power supply module is 100 W, but each
power supply is only delivering 50 W, the utilization factor of
each power supply is only 50 W/100 W or 50% utilization.
FIG. 2 is a block diagram illustrating another telecommunication
system 2B, which includes electronic device 10B coupled to AC power
sources (1, 2). Electronic device 10B is similar to electronic
device 10A; however, rather than including N+N power supply
modules, electronic device 10B includes N+1 power supply modules.
Also, telecommunication system 2B is similar to telecommunication
system 2A, but includes automatic transfer switch (ATS) 14. The
other components of electronic device 10B that are similar to
electronic device 10A are not described further.
As illustrated, ATS 14 couples to AC Source 1 and AC Source 2 and
selectively outputs the power from one of AC Source 1 and AC Source
2 (i.e., either AC source) to the N+1 power supply modules of
electronic device 10B. ATS 14 may reside at the system level (e.g.,
building power infrastructure level, and not at the equipment level
(i.e., within electronic device 10B), and switch between power
feeds for all of power supply modules.
In the example illustrated in FIG. 2, if AC Source 1 becomes
unavailable, ATS 14 switches power for device 10B from AC Source 1
to AC source 2, and the N+1 power supply modules deliver DC power
to cards 12 by receiving power from AC source 2. When power becomes
available again via AC Source 1, ATS 14 switches back from AC
Source 2 to AC Source 1, and the N+1 power supply modules deliver
DC power to cards 12 by receiving power from AC Source 1. For
instance, as can be seen ATS 14 is installed in AC distribution
line can supply AC power to the input of AC/DC PSMs either from AC
Source 1 or from AC Source 2. ATS 14 determines which of the AC
power sources provides the power at any time. ATS 14 control
detects loss of power or brownout condition for AC power source
delivering power at the time and transfers PSM input to second
redundant AC power source.
In the example illustrated in FIG. 2, there are N+1 power supply
modules where one power supply module (e.g., N+1.sup.th) functions
as a backup power supply module in case one of the other power
supply modules malfunctions. For example, during normal operation,
each of the N+1 power supply modules delivers DC power. In this
case, each power supply module delivers approximately (1/(N+1)) of
the total needed power, and if one of the power supply modules
malfunctions, then each power supply module delivers approximately
1/N of the total needed power.
There may be better power utilization in the example illustrated in
FIG. 2 as compared to the example illustrated in FIG. 1. For
example, in FIG. 1, each power supply module is configured to
deliver 1/N of the total needed power, and there are 2N power
supply modules resulting in approximately 50% utilization. In FIG.
2, each power supply module is also configured to deliver 1/N of
the total needed power, but there are N+1 power supply modules
resulting in approximately N/(N+1) utilization, which for all cases
expect where N equals 1 will result in better than 50% utilization.
Also, because there are N+1 power supply modules rather than 2N
power supply modules, there is a reduction in space needed to
include the power supply modules in an electronic device as well as
reduction in cost associated with power supply modules. In other
words, advantage of the scheme illustrated in FIG. 2 is reduced
number of PSM (N+1 vs. 2N) that results in better system cost
effectiveness, saving space occupied by PSMs, increased
reliability, and better PSM utilization factor (up to N/(N+1)% of
maximum PSM power vs. 50%).
However, there may be some drawbacks with the example illustrated
in FIG. 2. For example, standalone ATS 14 in the AC distribution
line introduces a single point of failure component. If for some
reason ATS 14 fails, the AC power to the entire telecommunication
system fails as well and the entire system operation is ceased.
For example, the relay in ATS 14 may become latched and remain
coupled to AC Source 1. In this case, if AC Source 1 becomes
unavailable (e.g., loss of power, loss of AC cycle, or brownout),
the relay in ATS 14 may not switch to AC Source 2, and electronic
device 10B may not receive power. As another example, circuitry
within ATS 14 that determines whether there is loss of power may
malfunction, and ATS 14 may not switch to the other power source in
response to a loss of power. In some cases, if ATS 14 completely
malfunctions, then even if both AC Source 1 and AC Source 2 are
capable of delivering power, ATS 14 may not couple the power to
electronic device 10B. In this sense, ATS 14 may create a single
point of failure without any redundancy for power delivery in the
event that ATS 14 fails or malfunctions.
FIG. 3 is a block diagram illustrating an example telecommunication
system 2C in which an example electronic device 10C is configured
to implement one or more of the techniques described in this
disclosure. For instance, in order to avoid single point of failure
component, the AC/DC PSM redundancy scheme N+1 with distributed
Automatic Transfer Switch (ATS) within each PSM is described in
this disclosure.
As illustrated in the example of FIG. 3, telecommunication system
2C includes electronic device 10C, which may be similar to
electronic device 10A and 10B. Electronic device 10C includes power
supply modules (PSMs) 18A-18N+1 (collectively referred to as PSMs
18). Each one of PSMs 18 includes a respective ATS 16A-16N+1
(collectively referred to as ATSs 16). An example of PSM 18 and ATS
16 is illustrated in greater detail in FIG. 4.
As illustrated in FIG. 3, the techniques described in this
disclosure may adopt some or all advantages of the scheme
illustrated in FIG. 2 and eliminate a single point of failure
component (e.g., ATS 14 of FIG. 2 installed in distribution line).
In FIG. 3, each one of PSMs 18 includes its own one of ATSs 16 that
allows keeping N+1 PSM redundancy and 1+1 AC source redundancy. In
FIG. 3, if for some reason one of ATSs 16 within one of PSMs 18
fatally fails, the system will still continue operating with N
pieces of PSMs 18 while the maintenance personnel can replace the
failed PSM.
Other example techniques have been described for switching from one
power source or another power source, such as those in U.S. Pat.
No. 5,939,799. However, these other techniques suffer from some
potential drawbacks.
In these other techniques, a determination of whether AC power has
been lost is made by monitoring the DC output of the power supply.
For instance, transfer power between two AC sources is actuated by
sensing of output voltage of AC/DC PSM. Such kind of ATS control
will not allow catching missing sine-wave cycles and AC source
brownout condition.
Furthermore, in these other techniques, a determination that AC
power has been lost may be made incorrectly. For example, a
component within the power supply module may malfunction causing
the voltage at the DC output to drop. In this case, the drop in the
DC voltage is from a malfunction of a component, and not from AC
power being lost. Nevertheless, these other techniques may still
determine that AC power is lost because DC voltage level dropped,
and unnecessarily cause the power supply module to switch from one
AC source to the other.
Also, the amount of time it takes the ATS to switch from one power
source to another may be too long (e.g., 100-200 milliseconds
(ms)). This requires backup battery/Uninterruptible Power Supply
(UPS) or another DC backup source connected to the load (e.g.,
cards 12) in order to maintain uninterruptable system operation.
Additional DC backup is expensive. Having such backup also
decreases reliability of entire system.
For example, the power supply modules, including PSMs 18, may
include a capacitor that charges during normal operation. When the
primary power feed (e.g., AC Source 1) becomes unavailable, and the
ATS is switching from the primary power feed to the secondary power
feed (e.g., AC Source 2), the capacitor may deliver the DC power to
the components of the electronic device so that the power to the
electronic device is uninterrupted. However, because the components
of the electronic device may require a relatively large amount of
power, the capacitor may only be able to deliver power for a
relatively short amount of time (e.g., in order of 40 ms). If the
time it takes the ATS to transfer from one source to another is in
the order of 100 ms to 200 ms, the size of the capacitor may be
impractically large and costly to serve the components for 100 ms
to 200 ms.
In these other techniques, the ATS may have a single mechanical
relay. The relay of the ATS may actuate any time it is determined
that the DC output voltage is low, including times when there is
high current flowing through the contacts. By switching when there
is high current flowing though the contacts, there may be
overvoltage spikes, which in turn decreases reliability of the
entire power system.
Furthermore, the ATS of these other techniques may include only one
relay in the current path. This introduces a potential hazard in
the event that the relay fails by shorting, which causes voltage
from one AC source to feed to the other AC source (e.g., voltage
from one AC source could be seen on input feed of second AC
source).
As described in more detail, the techniques described in this
disclosure may overcome one or more of the drawbacks described
above. In one example, this disclosure describes techniques for a
dual input uninterruptable AC/DC power supply for telecommunication
equipment with smart distributed automatic transfer switch (ATS).
As another example, the disclosure describes techniques that enable
enforcement of switching of one AC input to another when the former
is interrupted or faces a brownout condition. As another example,
the disclosure describes techniques that enable transferring power
from one AC source to another at the moment when input current
crosses zero. By ensuring that the input current is at zero, the
techniques may reduce arcing, overvoltage spikes, and increase life
time and reliability of ATS mechanical relays. The disclosure also
relates to supplying power to the PSM load (e.g., components of
electronic device 10C) within system during transfer process from
one power source to another, maintaining transfer time within
holdup time of the PSM, avoiding overcurrent condition, and making
uninterruptable operation of entire telecommunication system. In
telecommunication cases, this may result in uninterruptable
forwarding or processing of network traffic, as one example. The
disclosure also describes techniques that enable a device to meet
safety requirement in terms of having reinforced isolation between
two input feeds to avoid electrical hazard once AC feed is
disconnected.
FIG. 4 is a block diagram illustrating one example of a power
supply module that includes an automatic transfer switch (ATS) in
accordance with one or more examples described in this disclosure.
For example, FIG. 4 illustrates PSM 18A and ATS 16A in greater
detail. PSMs 18B-18N+1 and ATSs 16B-16N+1 may be substantially
similar to PSM 18A and ATS 16A illustrated in FIG. 4.
As shown in the illustrated example, PSM 18A is coupled to AC power
source 30 and AC power source 40. AC power source 30 may be the
primary power feed, and AC power source 40 may be secondary,
redundant power feed. AC power source 30 and AC power source 40 are
substantially similar to AC Source 1 and AC Source 2, respectively,
of FIGS. 1-3.
As illustrated, AC power source 30 is coupled to the L1 and N1
input terminals of PSM 18A, and AC power source 40 is coupled to L2
and N2 input terminals of PSM 18A. In the example of FIG. 4, L1 and
N1 form a first power feed, where L1 is the "hot" line and carries
the voltage, and N1 is the return, neutral line. L2 and N2 form a
second power feed, where L2 is the "hot" line, and N2 is the
return, neutral line. For purposes of brevity, L1 and N1 together
are referred to as a first power feed and L2 and N2 together are
referred to as a second power feed.
In this manner, PSM 18A is configured to accept a single phase AC
input from two independent AC sources (AC power source 30 and AC
power source 40). Although the example techniques are illustrated
with respect to a single phase AC input, the techniques described
in this disclosure may be extended to three-phase AC inputs, or
generally to N-phase AC inputs. For instance, in some examples, PSM
18A may be coupled to AC power source 30A and AC power source 40A,
PSM 18B may be coupled to AC power source 30B and AC power source
40B, and PSM 18C may be coupled to AC power source 30C and AC power
source 40C. In these examples, AC power sources 30A-30C form a
three-phase primary AC input, with each one of AC power sources
30A-30C representing a respective phase. AC power sources 40A-40C
form a three-phase secondary AC input, with each one of AC power
sources 40A-40C representing a respective phase.
In three-phase AC input examples, if one of PSMs 18A-18C switches
from a primary AC input to a secondary AC input, the other PSMs
18A-18C also switch from the primary AC input to a secondary AC
input, even if the primary AC input is still available. As an
example, if power source 30A becomes unavailable, PSM 18A in the
three-phase AC input example switches from AC power source 30A to
AC power source 40A. In this example, even if power sources 30B and
30C are available, PSM 18B and PSM 18C in the three-phase AC input
example switch from AC power source 30B and AC power source 30C to
AC power source 40B and AC power source 40C to maintain the
three-phase relationship.
In either example (i.e., single phase AC input or three-phase AC
input), each AC input of PSM 18A is connected to the input of
electromagnetic interference (EMI) filter 50 via contacts of relays
60-130, as illustrated in FIG. 4. In some examples, ATS 16A may be
considered as including relays 60-130, as illustrated by the dashed
line in FIG. 4. However, examples of ATS 16A are not so limited.
For instance, PSM 18A includes relay driver 280, and in some
examples, relay driver 280 may be part of ATS 16A. As another
example, PSM 18A includes micro-controller unit (MCU) 270. In some
examples, rather than one MCU 270, PSM 18A may include two
micro-controller units. In these examples, one of the two
micro-controller units may be part of ATS 16A. In this sense, ATS
16A within PSM 18A is illustrated conceptually to assist with
understanding the techniques described in this disclosure. The
example ATS 16A illustrated in FIG. 4 is not the only example of an
ATS, and PSM 18A may include other types of ATSs. For instance, PSM
18A may include an ATS such as one described in U.S. application
Ser. No. 13/946,916, filed Jul. 19, 2013, and entitled "AUTOMATIC
TRANSFER SWITCH SPACING MONITORING WITHIN AN ELECTRICAL DEVICE,"
the entire contents of which being incorporated by reference.
As one example, relays 60-130 may be the HF140FF012-2HWTF (456)
relay Xiamen Hongfa Electroacoustic Co., Ltd. However, the
techniques described in this disclosure are not limited to these
specific relays and other relays may be utilized.
In general, relays 60-130 may be relays with normally open contacts
(i.e., normally opened and need a voltage to close). One of the
specifications of relays 60-130 is the operate time, and another
specification of relays 60-130 is the release time. The relay
operate time is a time that begins when the voltage is applied to
the relay coil and ends when the relay wiper arm has reached
(without bouncing) the normally open contact. In other words, the
operate time is the time necessary to make the relay contact close.
The relay release time is a time that starts with the removal of
the voltage from the coil and ends when the wiper arm has returned
to the normally open contact. In other words, the release time is
the time necessary to open a closed relay.
In some examples, the operate time for relays 60-130 may be less
than or equal to 15 ms, and the release time for relays 60-130 may
be less than or equal to 5 ms. For example, in the worst case
scenario, relays 60-130 toggle from an open state to a close state
in 15 ms, and in the worst case scenario, relays 60-130 toggle from
a close state to an open state in 5 ms.
As an example, assume that relays 60, 70, 100, and 110 are in a
close state and relays 80, 90, 120, and 130 are in an open state.
In this example, assume that relay driver 280 outputs a signal that
causes relays 60, 70, 100, and 110 to open and causes relays 80,
90, 120, and 130 to close. In this case, it may take a maximum of 5
ms for relays 60, 70, 100, and 110 to toggle to the open state, and
it may take a maximum of 10 more ms after the 5 ms (i.e., for a
total of 15 ms) for relays 80, 90, 120, and 130 to close.
As described in more detail below, in some examples, relay driver
280 includes circuitry that reduces the operate time from less than
or equal to 15 ms to less than or equal to 5 ms. In this manner,
the techniques further reduce the amount of time it takes PSM 18A
to transition from one power feed to another (e.g., the operate
time is reduced by 10 ms), which in turn provides additional buffer
for the time it takes to ensure that the input current reached zero
before the transition from one power feed to another occurs.
The output of EMI filter 50 is connected to the input of diode
bridge rectifier 400 that includes diodes 140-170 and
silicone-controller rectifiers (SCRs) 290 and 300. As described in
more detail, diodes 140 and 150 may initially be part of the
conducting circuitry of the diode bridge rectifier, and SCRs 290
and 300 may be de-asserted. However, after PSM 18A reaches steady
state (normal operating conditions), diodes 140 and 150 may be
decoupled from the conducting circuitry and be replaced by SCRs 290
and 300, respectively (i.e., SCRs 290 and 300 are asserted and form
part of the diode bridge rectifier).
The function of the diode bridge rectifier 400 is to convert the
negative portions of the AC input to a positive component such that
the rectified voltage includes only positive portions. As described
in more detail, by asserting and de-asserting SCRs 290 and 300, the
techniques described in this disclosure may ensure that PSM 18A
transitions from AC power source 30 to AC power source 40, in the
event that AC power source 30 becomes interrupted, when the input
current has reached approximately zero amperes (Amps or A). For
instance, if AC power source 30 becomes interrupted (e.g.,
unavailable, misses an AC cycle, or due to a brownout), MCU 270 may
de-assert SCRs 290 and 300. However, even though SCRs 290 and 300
are de-asserted, SCRs 290 and 300 may still conduct some current.
In other words, SCRs 290 and 300 conduct current until the current
becomes substantially zero even when SCRs 290 and 300 are
de-asserted after being asserted.
After AC power source 30 becomes interrupted, there may be some
transient current that is still flowing through ATS 16A. In some
examples, there may be a threshold amount of time before such a
transient current to reaches zero (e.g., 15 ms as a worst-case
scenario). In the techniques described in this disclosure, after
MCU 270 determines that AC power source 30 is interrupted, MCU 270
may de-assert SCRs 290 and 300. In this case, although SCRs 290 and
300 are de-asserted, SCRs 290 and 300 may conduct any remaining
transient current. The techniques described in this disclosure may
wait a threshold amount of time (e.g., 15 ms) and then cause relays
60, 70, 100, and 110 to toggle to an open state, and cause relays
80, 90, 120, and 130 to toggle to a close state so that PSM 18A
receives power from power source 40.
For example, MCU 270 may output a signal to SCR control unit 350
that instructs SCR control unit 350 to de-assert the gate voltage
on SCRs 290, 300. At the time that MCU 270 instructs SCR control
unit 350 to de-assert the gate voltage on SCRs 290, 300, MCU 270
may start a counter that counts to 15 ms. After the counter reaches
15 ms, MCU 270 may output a signal to relay driver 280 that
instructs relay driver 280 to cause relays 60, 70, 100, and 110 to
toggle to an open state, and cause relays 80, 90, 120, and 130 to
toggle to a close state. In some examples, the time it takes relays
60, 70, 100, and 110 to toggle to an open state may be a maximum of
5 ms (the release time) and the time it takes relays 80, 90, 120,
and 130 to toggle to a close state may also be a maximum of 5 ms
(the operate time) using the techniques described in this
disclosure.
In this manner, relays 60, 70, 100, and 110 may open and relays 80,
90, 120, and 130 may close after the input current is guaranteed to
have reached zero (e.g., after any remaining transient current has
been conducted). By ensuring the input current has reached zero
before opening relays coupled to power source 30 and closing relays
coupled to power source 40, the techniques may minimize the chances
of arching and overvoltage spikes.
As illustrated in FIG. 4, the output of bridge rectifier 400 is
connected to the input of active PFC (Power Factor Correction)
block 180. PFC 180 converts the rectified AC voltage from bridge
rectifier 400 into a DC voltage. For example, PFC 180 may include a
capacitor, which is also a bulk capacitor for providing power to
the components of the electronic device during transition from one
power source to another. The output of PFC 180 is connected to the
input of DC/DC converter 190 which generate DC voltage Vout at its
output terminals OUT_P and OUT_N that further is distributed to the
system components (e.g., components of electronic device 10C such
as one or more cards 12). OUT_P may form the positive voltage and
OUT_N may form the return ground.
In order to transfer power between two AC power sources 30 and 40
at the moment when AC sine-wave voltage cycles is missing or AC
source is in brownout state, and also in order to switch back to
the AC power source in a case where its power is restored, PSM 18A
includes input voltage sensors 220 and 230 to sense voltage across
L1/N1 and L2/N2, respectively, and current sensors 240 and 250 to
sense current outputted via L1 and L2, respectively. Although both
voltage sensors and current sensors are illustrated, the techniques
described in this disclosure may not require both voltage and
current sensors, and may function with either voltage sensors or
current sensors.
Voltage sensor 220 and current sensor 240 may be configured to
output a signal indicative of an AC voltage or an AC current,
respectively, at a connection of power supply source 30 to PSM 18A
to MCU 270. MCU 270 may receive the signal indicative of the AC
voltage or AC current and determine whether power from power source
30 is unavailable. For example, MCU 270 may determine that power
from power source 30 is unavailable when one or both of the AC
voltage or AC current is less than a threshold. MCU 270 may
similarly determine whether power from power source 40 is
unavailable based on signals received from voltage sensor 230 and
current sensor 250. For example, MCU 270 may receive the outputs of
all input voltage and current sensors.
One of the outputs of MCU 270 is connected to the relay driver unit
280 to switch on/off respective relays 60-130. In some examples,
rather than being implemented separately, one or more of relays
60-130 may be implemented together in a common set of relays. For
example, a first set of relays may include both relay 60 and relay
80, a second set of relays may include both relay 70 and relay 90,
a third set of relays may include both relay 100 and 120, and a
fourth set of relays may include both relay 110 and 130. In this
example, the first, second, third, and fourth sets of relays may be
configured such that when the relays receive a signal from relay
driver unit 280 one of the relays closes and the other opens in
response to the single signal from delay driver unit 280. As one
example, if the first set of relays received a digital high, the
first set of relays may close relay 60 and open relay 80, and if
the first set of relays received a digital low, the first set of
relays may open relay 60 and close relay 80. In this way, relays
60-130 may open and close as needed with a single output from relay
driver unit 280, rather than two different signals (one for opening
or closing relays 60, 70, 100, and 110, and other signal for
opening or closing relays 80, 90, 120, and 130), which reduces the
chances of relays 60-130 not opening or closing as needed. For
example, the sets of relays coupled to two AC inputs (e.g., AC
power source 30 via L1/N1 and AC power source 40 via L2/N2) may be
driven from one coil, which may reduce or eliminate the possibility
of incorrect relay switchover shorting two sources.
Also, ATS 16A is illustrated as including relays 60 and 70 in
series and relays 100 and 110 in series for AC power source 30, and
relays 80 and 90 in series and relays 120 and 130 in series for AC
power source 40. Such series connection further reduces the chances
of creating a current path between AC power source 30 and AC power
source 40. For example, if there was only one relay (rather than
two in series) and this single relay malfunctions, there may be
current path via arching between AC power source 30 and AC power
source 40. However, the chances that both relays in series will
malfunction simultaneously may be much lower than the chance that a
single relay will malfunction. Accordingly, by connecting the
relays in series within ATS 16A, the chances of AC power source 30
and AC power source 40 coupling may be minimized. In other words,
in order to meet safety requirement for telecommunication equipment
and rule out major safety hazard in case if relay fails shorted
each input AC line connection has two relays connected in series.
For instance, two relays in series in each input may avoid safety
hazard in case one relay fails. The pairs of these relays are as
follow: 60-70, 80-90, 100-110, and 120-130.
The techniques described in this disclosure may keep the transfer
time (e.g., time to transition from one power feed to another)
within a holdup time of the PSM (e.g., less than 40 ms). By keeping
the transfer time relatively short (e.g., 40 ms), the techniques
allow for a smaller sized bulk capacitor in PFC 180 as compared to
the capacitor needed for a longer transfer time. For example, as
described above, the capacitor in PFC 180 may deliver power to
components of electronic device 10C during the transfer from one
power source to another. The size of the capacitor may be
proportional to the transfer time (e.g., the longer the transfer
time the larger the capacitor, the shorter the transfer time the
smaller the capacitor). A smaller sized capacitor is generally
preferable because the smaller sized capacitor requires smaller
space allowing for smaller sized PSMs 18 and because the smaller
sized capacitor costs less.
In the techniques described in this disclosure, in order to avoid
arcing at relay contacts, to avoid overvoltage spikes, and to
increase reliability of mechanical relays 60-130 during
transferring process, PSM 18A includes two SCRs 290, 300, auxiliary
relay 310, with series resistor 320, voltage sensor 330, current
sensor 340, and SCR control unit 350. Voltage sensor 330 and
current sensor 340 sense voltage and a current at the input of
diode rectifier bridge 140-170. Output voltage sensor 360 is to
sense the Vout (e.g., voltage across OUT_P and OUT_N) and provide
this information to the MCU 270. Voltage sensor 330, 360 and
current sensor 340 may not be needed in every example and are
provided for purposes of illustration only.
In some examples, MCU 270 may receive a signal indicative of the
voltage from the voltage sensor 360 and determine when to
transition from power source 30 to power source 40. For example, as
described above, when power source 30 becomes unavailable, the bulk
capacitor in PFC 180 delivers power to the components of electronic
device 10C. The length of time that the bulk capacitor can deliver
power is a function of the output load. For instance, if cards 12
do not require much power, the bulk capacitor can deliver power for
longer than if cards 12 required a great amount of power.
As the bulk capacitor delivers power, the voltage at the output of
PSM 18A (e.g., across OUT_P, OUT_N) begins to droop down. The time
when the voltage at the output of PSM 18A begins to droop and the
rate at which the voltage at the output of PSM 18A droops may be a
function of the amount of power that the one or more components
need. In some examples, MCU 270 may determine based on the signal
from voltage sensor 360 the rate of the droop on the output voltage
and the level of the output voltage.
If the rate of the droop on the output voltage is relatively slow,
MCU 270 may determine that transitioning from power source 30 to
power source 40 can be delayed because the bulk capacitor is
capable of providing power for a longer time. In this case, if
power from power source 30 is restored during the additional delay
in transitioning, no transitioning from power source 30 to power
source 40 is needed, which extends the lifetime of ATSs 16. If the
rate of the droop on the output voltage is relatively fast, MCU 270
may determine that transitioning from power source 30 to power
source 40 may need to be done earlier because the bulk capacitor
may not be capable of providing power for much longer.
Examples of MCU 270 include, but are not limited to, one or more
digital signal processors (DSPs), general purpose microcontrollers,
application specific integrated circuits (ASICs), field
programmable logic arrays (FPGAs), or other equivalent integrated
or discrete logic circuitry. In some examples, MCU 270 may receive
power from AC source 30 and AC source 40 or from the output of
DC/DC converter 190. In some examples, MCU 270 of PSM 18A may also
receive power from the Vouts of other PSMs 18. In this manner, if
there is a fatal defect in the electrical components of PSM 18A,
MCU 270 may still receive power from other functioning PSMs 18.
In some examples, MCU 270 may be configured to record events within
PSM 18A. As one example, MCU 270 may be configured to record the
number of times the relays of ATS 16A toggled. As another example,
MCU 270 may record the number of times AC power source 30 or AC
power source 40 became unavailable. MCU 270 may record other such
information that is generally indicative of the operational
characteristics of PSM 18A. MCU 270 may output such information to
electronic device 10C (e.g., one or more processors of electronic
device 10C). Electronic device 10C may assert an alarm if the
received information indicates a potential error or that a
component is operating at below specification levels in PSM 18A.
For example, if the received information indicates that the relays
are close to the maximum number of toggle times, electronic device
10C may assert an alarm indicating that PSM 18A should be
replaced.
In the example illustrated in FIG. 4, the output of both voltage
and current sensors connected to the respective inputs of MCU 270
while one of the MCU 270 output is connected to the input of SCR
control unit 350. The output of SCR control unit 350 is connected
to the gates of SCR 290, 300 and a coil of relay 310.
FIG. 5 is a timing diagram illustrating techniques in accordance
with one or more examples described in this disclosure. FIG. 6 is a
flowchart illustrating the corresponding timing in FIG. 5. FIGS. 5
and 6 should be consulted together as FIG. 6 provides additional
context for the timing diagram of FIG. 5. For purposes of
illustration, the techniques are described with respect to FIG.
4.
Assume that initially AC power source 30 is connected to the input
L1/N1 of PSM 18A while AC power source 40 is connected to the input
L2/N2 of PSM 18A. At start up cycle of PSM 18A, at time T0, MCU 270
enables voltage to the coils of respective relays to prepare PSM
18A being connected to the default input. In this example L1/N1 of
AC power source 30 is consider as default one. For instance, at
time T0, MCU 270 enables voltage to the coil of relays 60, 70, 100,
110 to close relays 60, 70, 100, and 110. At time T1 (e.g.,
approximately 5 ms after time T0), the contacts of relays 60, 70,
100, 110 close and AC voltage from L1/N1 is connected to the input
of EMI filter 50. After that PSM 18A starts operating as any
conventional AC/DC PSM. For instance, at time T1, Vout voltage on
output OUT_P, OUT_N ramps up as the capacitor within PFC 180
charges up.
Auxiliary relay 310 may be relay with normally closed contacts,
which means that if no voltage is applied to relay 310, relay 310
connects resistor 320 to diode 140 and 150. At time T1, while the
Vout voltage is ramping up, MCU 270 causes SCR control unit 350 to
not assert a voltage to relay 310 keeping resistor 320 connected in
the current path. Also, at time T1, SCR control unit 350 may not
assert a voltage to SCRs 290, 300 keeping SCRs 290, 300 from
conducting current.
Series resistor 320 limits PSM 18A inrush current. For example, at
start-up there may be rush of current into PSM 18A, and resistor
320 limits such an inrush current. Also, during this time, all
voltage and current sensors monitor respective voltages and
current. At time T2, the output voltage sensor 360 may signal to
MCU 270 that transition process of ramping up of Vout is completed
and Vout reached its stable value. MCU 270 may generate control
signal to the SCR control unit 350 that in turn asserts control
signal on its outputs to turn SCRs 290, 300 on (e.g., assert a
voltage on the gates of SCRs 290, 300). At time T3, SCR control
unit 350 asserts signal to the coil of auxiliary relay 310 to cause
relay 310 to open.
At time T4, relay 310 opens. Accordingly, at time T4, SCRs 290, 300
are on and series resistor 320 is no longer connected (i.e.,
bypassed). When series resistor 320 is not connected and SCRs 290,
300 are on, SCRs 290, 300 bypass diodes 140, 150. In other words,
open contact of relay 310 takes resistor 320 and diodes 140, 150
out of the current (i.e., conducting) path. The PSM 18A now
operates in normal condition from AC source 30 as long as voltage
sensor 220 senses proper voltage on L1/N1 (e.g., provides its
N+1.sup.th amount of power for the components of electronic device
10C).
At time T5 (which is at some time later), AC input L1/N1 is lost
(either missing cycle or brownout condition for AC power source
30). Voltage sensor 220 on L1/N1 senses this condition and provides
this information to MCU 270. Because AC input L1/N1 is lost, Vout
voltage starts decaying down. During this time PSM 18A continues
supplying energy to the output converting energy stored in bulk
capacitor on output of PFC 180 inside of PSM 18A. Because PFC
voltage is about 380-400 VDC this capacitor stores enough energy to
supply output voltage during necessary holdup time. As the
capacitor discharges, the Vout voltage starts decaying down (e.g.,
droops), as illustrated by the decaying on the OUT_P, OUT_N line in
FIG. 5.
As described above, the rate at which the voltage at the output of
PSM 18A (i.e., Vout) begins to droop (i.e., decay) may be a
function of the amount of power that the one or more components
need. In other words, the Vout voltage decaying down slew rate
depends on PSM 18A external load. As one example, if there is no
heavy network traffic at the time, the PSM load may be relatively
light. Accordingly, the amount of time that the bulk capacitor can
provide power may be variable. The amount of time that the bulk
capacitor can deliver power may be referred to as a ride through
period because the ride through period defines how long power to
electronic device 10C can be delivered before back up power is
needed. For instance, the ride through period defines a tolerable
delay after power from the primary power source becomes unavailable
and power from the backup power source is needed.
If the slew rate of the Vout voltage decaying relatively slow, then
the ride through period may be longer than if the Vout voltage
decaying slew rate is relative fast. In some examples, MCU 270 may
continuously calculate the ride through period after power from
power source 30 is interrupted to determine when to begin the
transition from power source 30 to power source 40. The ride
through period may also be referred to as a holdup time
(Tholdup).
MCU 270 may utilize the following equation to determine the ride
through period (i.e., the holdup time or Tholdup):
.eta. ##EQU00001##
In the above equation, Tholdup is the ride through period (i.e.,
the amount of time after time T5 that MCU 270 can wait before
transitioning from power source 30 to power source 40). In other
words, MCU 270 may transition from power source 30 to power source
40 based on the determined ride through period. The variable .eta.
is PSM efficiency, Cbulk is the value of the bulk capacitor of PFC
180, Vout is the nominal value of the output voltage, Vdropout is
the minimum value of the output voltage Vout that allows electronic
device 10C to operate properly. Pout is the PSM actual output power
drawn at the time.
In the above equation, the holdup time (i.e., ride through period
or Tholdup) is inversely proportional to Pout. Accordingly, when
power from power source 30 is unavailable at time T5, MCU 270
determines the expected holdup time using the above equation. MCU
270 may continuously determine the expected holdup time and make
corrections of the starting point time of when to transition from
power source 30 to power source 40.
In some examples, instead of determining the ride through period on
the fly (i.e., by continuously calculating the holdup time), MCU
270 may be preconfigured with a ride through period (e.g.,
preconfigured with a ride through period of 16 ms). In the case of
a predetermined ride through period, although MCU 270 may not need
to continuously determine the ride through period, there may be
some drawbacks. For example, with a predetermined ride through
period, MCU 270 may transition from power source 30 to power source
40 too quickly (i.e., there may be sufficient power in the bulk
capacitor to keep delivering power when MCU 270 transitions from
power source 30 to power source 40). In some examples, power source
30 may become available, but MCU 270 may have already and
unnecessarily transitioned to power source 40. Utilizing the above
equation to determine the ride through period on the fly, the
techniques may drastically reduce the number of unnecessary
transfer events, which results in increase reliability of the ATS
and the entire PSM.
After the ride through period (e.g., as determined by MCU 270 or
predetermined), at time T6, MCU 270 asserts signal to SCR control
unit 350 to cause SCR control unit 350 to de-assert the gate
voltage on SCRs 290, 300 causing SCRs 290, 300 to turn off. In some
examples, MCU 270 may determine the ride through period to avoid
transitioning from AC power source 30 to AC power source 40 in case
power via AC power source 30 returns. For example, AC power source
30 may be interrupted for a brief time, and switching over may not
be needed.
At time T6, after MCU 270 and SCR control unit 350 de-assert SCR
290, 300 (e.g., with control signals to the gates of SCRs 290,
300), current may still continue flowing through SCRs 290, 300, as
this is a characteristic of SCRs. Because SCRs 290, 300 gate signal
is no longer active, input sine wave current flowing through either
SCR 290 or SCR 300 will be interrupted once the current crosses
zero point. At this moment SCRs 290, 300 are in open state.
In other words, there is a holding current level associated with
SCRs 290, 300. Even if MCU 270 and SCR control unit 350 de-assert
SCRs 290, 300, SCRs 290, 300 still conduct current until the
current through SCRs 290, 300 falls below the holding current
level. In some examples, it may take up to 15 ms from the time that
SCR control unit 350 de-asserts SCRs 290, 300 for the current
through SCRs 290, 300 to reach zero (fall below the holding current
level). For instance, it may take up to 15 ms from the time that
SCRs 290, 300 are turned off for the input current to reach zero,
but the actual amount of time it takes the input current to reach
zero may be random (but less than 15 ms).
At time T7 (e.g., 15 millisecond later of T6 to guarantee that the
input current reached zero), MCU 270 may output a signal to relay
driver 280 causing relay driver 280 to de-energize coils of relays
60, 70, 100, and 110 (e.g., cause relays 60, 70, 100, and 110 to go
into an open state) and energize coils of relay s80, 90, 120, and
130 (e.g., cause relays 80, 90, 120, and 130 to go into a close
state). For example, at time T6, MCU 270 may start a counter, and
when the counter indicates that 15 ms have elapsed, MCU 270 may
output the signal to relay driver 280 to cause relays 60, 70, 100,
and 110 to open and cause relays 80, 90, 120, and 130 to close.
Accordingly, the contacts of relays 60, 70, 100, 110 go from close
state to open state when current through these contacts is zero
because SCR 290 and SCR 300 were turned off early and interrupted
input current. This allows avoiding relay contacts arcing, increase
relay reliability, and avoid input overvoltage spikes.
At time T8 (e.g., 5 millisecond later) contacts of relay 60, 70,
100, and 110 are open disconnecting PSM 18A from L1/N1 (e.g., from
AC power source 30), and contacts of relays 80, 90, 120, 130 are
close connecting PSM 18A to L2/N2 (e.g., to AC power source 40).
Thus, PSM 18A again is connected to the AC source. At time T9 when
voltage sensor 230 coupled to L2/N2 detects right value of voltage
from AC power source 40 (e.g., approximately 2 ms after time T8),
MCU 270 and SCR control unit 350 asserts signal to the gates of
SCRs 290, 300 turning them on creating path for the input current
from AC power source 40. This causes PSM output voltage (Vout) to
start ramping up to its normal operating level (i.e., nominal
value).
In this example, assuming 16 ms for the determined ride through
period, the total transfer time from AC source 30 to AC source 40
is about 38 milliseconds (e.g., 16 ms for ride through period plus
15 ms to ensure current through SCRs reach zero plus 5 ms to toggle
the relays plus 2 ms for detecting power at L2/N2 is correct).
During this transfer time bulk capacitor on the output of PFC 180
provides energy to the output to keep the output voltage within
specified range in order to maintain uninterruptable operation of
entire telecommunication system. At time T10 the output voltage
reached its nominal value and PSM 18A continues normal operation
from AC power source 40 until the voltage from failed AC power
source 30 is restored to its nominal value.
At some time T11, when voltage from AC power source 30 is restored
to its nominal condition ATS 16A initiates transfer back from AC
power source 40 to the default, main AC power source 30 (i.e.,
transitions from the secondary power feed to the primary power
feed). The transfer back processes are similar to the above
described process for transferring to the secondary power feed. At
time T11, input voltage sensor 220 starts sensing proper voltage on
L1/N1, and signals information indicating that the voltage across
L1/N1 is at the correct level to MCU 270. In some examples, in
order to make sure that voltage on L1/N1 is indeed restored, MCU
270 waits for a restore period (e.g., approximately 5 seconds) and
then along with relay driver unit 280 at time T12 (e.g.,
approximately 5 seconds after T11) de-asserts SCRs 290, 300 to
begin the transfer back process.
Waiting for the restore period is not necessary in every example.
The purpose for waiting for the restore period is to not switch
unnecessarily until there is some assurance that AC power source 30
is fully functioning. For instance, if MCU 270 immediately
transitioned from AC power source 40 back to AC power source 30,
and AC power source 30 was not fully restored, then MCU 270 may
immediately transition from AC power source 30 back to AC power
source 40. Such a process may potentially repeat multiple times
causing chatter. By waiting for the restore period, the techniques
may minimize such chatter. Moreover, the number of times the relays
can toggle from one position to another may be fixed. For instance,
relays 60, 70, 100, and 110, and 80, 90, 120, and 130 may be
mechanical relays, and repetitive toggling of the mechanical relays
may speed the end of life of the mechanical relays.
At time T13, which approximately 15 ms after time T14 to ensure
that the input current is at zero, MCU 270 causes relay driver 280
to de-energize the coils of relays 80, 90, 120, and 130 (cause
relays 80, 90, 120, and 130 to open) and energize the coils of
relays 60, 70, 100, and 110 (cause relays 60, 70, 100, and 110 to
close). At time T14, approximately 5 ms after time T13, relays 80,
90, 120, and 130 open and relays 60, 70, 100, and 110 close.
At time T15 voltage sensor 230 confirms that input L1/N1 is correct
and MCU 270 along with SCR control unit 350 assert on voltage to
SCR 290, 300 gates turning them on approximately 2 ms after time
T14. The output of PSM 18A starts ramping up again and roughly
reaches its nominal steady value in 40 millisecond. At time T16
Vout voltage is correct and PSM 18A operates again normally from
main default AC power source 30 over input L1/N1.
In this manner, the techniques described in this disclosure are
related to an N+1 PSM and 1+1 source redundancy system without any
single point of failure. Also, the transition from one power source
to another occurs during zero current through the relays (i.e.,
zero current relay transfer).
Moreover, in case of transfer to backup power, the PSM may not
transfer back to the primary input for a predetermined time (e.g.,
restore period such as 5 seconds) unless the backup input goes
down, which reduces chatter and increases reliability. In some
examples, MCU 270 may determine a ride through period, and wait for
the determined ride through period before starting the source
switch over.
In some examples, the techniques may drive one set of relays in two
AC inputs from one coil. This may eliminate or reduce possibility
of incorrect relay switchover shorting two sources.
In some examples, the MCUs of respective PSMs 18 may also be
powered by the bias from common output voltage bus from the system
(e.g., the MCU of one of PSMs 18 may receive power from another one
or more of PSMs 18). In this manner, status information, such as
ATS status communication, to the system may never go down even if
there is no AC input to the PSM. In some examples, ATS transfer
events may be recorded and communicated to the system (a processor
of electronic device 10C). The system may raise alarm to get a PSM
replace if the events get close to the maximum transfer
specification of the relays.
In the above techniques, to ensure that transition from a first
power source to a second power source occurs when input current is
low (e.g., approximately zero), SCRs 290, 300 de-couple components
of electronic device 10C from receiving power from the first power
source prior to ATS 16A de-coupling PSM 18A from receiving power
from the power source. For example, during normal operation, one or
more components of electronic device 10C receive DC power converted
from AC power of power source 30. Then, when the power from power
source 30 is lost, MCU 270 de-assert SCRs 290, 300 prior to
toggling relays of ATS 16A. Therefore, prior to relays of ATS 16A
de-coupling PSM 18A from power source 30, SCRs 290, 300 de-couple
components of electronic device 10C from receiving power from power
source 30. Such de-coupling of the components of electronic device
10C from power source 30 ensures that the input current (i.e.,
current flowing into PSM 18A) is given sufficient time to fall to a
low amplitude (e.g., approximately OA). Then, when the input
current is approximately zero, MCU 270 toggles the relays of ATS
16A to de-couple PSM 18A from power source 30 and couple PSM 18A to
power source 40.
In this sense, SCRs 290, 300 may be considered as de-coupling
components (e.g., de-coupling components of a bridge rectifier)
that de-couple a first power source from components of an
electronic device before an ATS de-couples a power supply module
from the first power source. However, SCRs 290, 300 are not the
only example of coupling components, and other examples of coupling
components may be utilized. Furthermore, these coupling components
need not necessarily reside in a bridge rectifier, and the
inclusion of SCRs 290, 300 in bridge rectifier 400 is provided for
purposes of illustration only. Also, PSMs 18 need not necessarily
include two coupling components, and may instead include one or
more coupling components that de-couple a first power source from
components of an electronic device before an ATS de-couples a power
supply module from the first power source.
In other words, in the techniques described in this disclosure, one
or more components (e.g., cards 12) of electronic device 10C are
connected to an output of PSM 18A (e.g., OUT_P, OUT_N of PSM 18A
via the common Vout bus). The one or more decoupling components
(e.g., SCRs 290, 300) of PSM 18A connect ATS 16A to the output of
PSM 18A. For example, in normal operation, for current to flow from
the ATS 16A to the output of PSM 18A (OUT_P, OUT_N), the current
flows through SCRs 290, 300. In this way, the one or more
decoupling components may be considered as connecting ATS 16A of
PSM 18A to the output of PSM 18A.
Accordingly, in the techniques described in this disclosure, MCU
270 of PSM 18A may determine whether power from power source 30 is
unavailable to PSM 18A (e.g., based the signal received from AC
voltage sensor 220 and/or AC current sensor 240). Responsive to
determining that power from power source 30 is unavailable, MCU 270
may cause the one or more de-coupling components (e.g., SCRs 290,
300) to de-couple the one or more components of electronic device
10C that are connected to the output of PSM 18A (e.g., cards 12)
from power source 30. For example, the one or more de-coupling
components may stop receiving power from power source 30.
For instance, in normal operation, a rectified version of the AC
output of power source 30 charges the bulk capacitor that delivers
power to one or more components of electronic device 10C that are
connected to the output of PSM 18A. In this way, the one or more
components of electronic device 10C indirectly receive power from
power source 30. In the techniques described in this disclosure,
when power from power source 30 becomes unavailable, the one or
more de-coupling components de-couple the one or more components of
electronic device 10C connected to the output of PSM 18A from
receiving power from power source 30.
Subsequent to causing the one or more de-coupling components to
de-couple power source 30 from the one or more components of
electronic device 10C, MCU 270 may cause ATS 16A to de-couple PSM
18A from power source 30. For example, MCU 270 may wait until the
current through the de-coupling components (e.g., SCRs 290, 300) is
approximately zero before causing ATS 16A to de-couple PSM 18A from
power source 30. MCU 270 may cause ATS 16A to couple PSM 18A to
power source 40 for delivering power to the one or more components
of electronic device 10C.
FIG. 7A is a block diagram illustrating an example of a relay
driver of a power supply module. For instance, FIG. 7A illustrates
an example of relay driver 280 of PSM 18A of FIG. 4 in greater
detail. Relay driver 280 may drive relays 60, 70, 100, and 110, and
relays 80, 90, 120, and 130. For purposes of illustration, relay
driver 280, in FIG. 7A, is illustrated as driving one of the relays
of ATS 16A. For instance, output 608 of relay driver 280, in FIG.
7A, may drive relay 60. Relay driver 280 may include similar
circuitry for driving the other relays as well. For purposes of
illustration and brevity, the example illustrated in FIG. 7A is
described with respect relay driver 280 driving relay 60.
As described above, the operate time (time it takes to close) for
relay 60 may usually be less than or equal to 15 ms and the release
time (time it takes to open) for relay 60 may usually be less than
or equal to 5 ms. With the circuitry of relay driver 280,
illustrated in FIG. 7A, the operate time for relay 60 may be
reduced to less than or equal to 5 ms, and the release time may
remain at less than or equal to 5 ms.
Relay 60 may be configured to operate at a first voltage level
(e.g., 12V). For instance, when 12V is applied to relay 60, relay
60 toggles from the normally open state to the close state, and
when 12V is removed from relay 60, relay 60 toggles from the close
state to the open state. When relay 60 operates at the first
voltage level, the operate time and release time for relay 60 may
be specified operate time and release time (e.g., .ltoreq.15 ms for
the operate time and .ltoreq.5 ms for the release time).
If relay driver 280 applies a second voltage level greater than the
first voltage level to relay 60, then the operate time for relay 60
may be reduced. For example, if relay driver 280 applies 24V to
relay 60, then relay 60 may toggle from the open state to the close
state in less than or equal to 5 ms (e.g., the operate time is
reduced from worst case 15 ms to worst case 5 ms). However, it may
undesirable to apply the second voltage level to relay 60 for an
extended period of time because applying the second voltage level
to relay 60 may cause relay 60 to burn out.
Accordingly, the circuitry of relay driver 280, illustrated in FIG.
7A, may apply a second voltage level to relay 60 for brief instance
and then revert back to applying a first voltage level to relay 60
so that the operate time of relay 60 is reduced, but the chances of
damaging relay 60 is reduced. For example, when MCU 270 determines
that relay 60 should toggle to the close state, relay driver 280
may apply a first voltage level to voltage input 612 (e.g., apply
12V to input 612). In addition, relay driver 280 or MCU 270 may
apply a voltage pulse to relay speed input 614. The pulse width of
the voltage pulse may be in the order of milliseconds (e.g.,
approximately 10 ms). Voltage input 610 may be coupled to a second
voltage level (e.g., 24V) permanently or whenever relay driver 280
applies the first voltage level to voltage input 612 or whenever
relay driver 280 or MCU 270 apply a voltage pulse to relay speed
input 614.
In normal operation when no pulse is applied to relay speed input
614, transistors 600, 602, 604, and 606 may be off. Because
transistor 606 is off, transistor 606 blocks the second voltage
level (e.g., 24V at voltage input 610) from being applied to output
608. Then, when it is time to toggle relay 60 to the close state,
relay driver 280 applies the first voltage level (e.g., 12V) to
voltage input 612 and relay driver 280 or MCU 270 applies a voltage
pulse to relay speed input 614.
During the on-time of the voltage pulse (e.g., the 10 ms pulse
width), the voltage pulse turns on transistor 600, which turns on
transistor 602, which turns on transistor 604. Turning on
transistor 604 turns on transistor 606, and the second voltage
level at voltage input 610 applies voltage to relay 60 via output
608 (e.g., applies 24V to relay 60 via output 608). Diode 618
blocks current from flowing from voltage input 610 to voltage input
612. In this case, relay 60 toggles from the open state to the
close state in less than or equal to 5 ms because 24V is applied to
relay 60.
After the on-time of the voltage pulse (e.g., when the voltage at
relay speed input 614 is approximately zero), transistor 600 turns
off, which causes transistor 602 to turn off, which causes
transistor 604 to turn off, which then causes transistor 606 to
turn off. When transistor 606 is off, the second voltage level at
voltage input 610 is blocked from being applied to relay 60, and
instead the first voltage level at voltage input 612 is applied to
relay 60.
Accordingly, relay driver 280 may briefly apply a voltage that
causes the relays to transition from the open state to a close
state, and then apply a lower voltage that causes the relays to
stay in the close state. In this manner, the operate time of relay
60 may be reduced, but the chances of damaging relay 60 is limited.
Also, when relay 60 is to return back to the open state, relay
driver 280 may remove the voltage at voltage input 612. Therefore,
there may be no effect on the release time of relay 60.
FIG. 7B is a block diagram illustrating another example of a relay
driver of a power supply module. FIG. 7B is similar to FIG. 7A, and
similar components are referenced with the same numerical
references. FIG. 7B further provides a timing diagram which further
illustrates the timing at time T7 to illustrate how fast the coils
energize and de-energized.
FIG. 8 is a flowchart illustrating example operation of a power
supply module in accordance with one or more techniques of this
disclosure. For purposes of illustration, reference is made to
micro-controller 270 of PSM 18A of FIG. 4.
As illustrated in this example, a controller (e.g.,
micro-controller 270) of a power supply module (e.g., PSM 18A of
electronic device 10C) may determine whether power from a first
power source (e.g., AC power source 30) is unavailable to the power
supply module (700). For example, at least one of voltage sensor
220 and current sensor 240 may measure an AC voltage or an AC
current, respectively, at a connection of AC power source 30 to PSM
18A. Micro-controller 270 may receive a signal indicative of the AC
voltage or AC current and determine that power from AC power source
30 coupled to PSM 18A is unavailable when one or both of the AC
voltage and AC current is less than a threshold.
The controller may be configured to cause one or more de-coupling
components of the power supply module that connect an automatic
transfer switch (ATS) of the power supply module to an output of
the power supply module to de-couple the first power source from
the one or more components of the electronic device that are
connected to the output of the power supply module in response to
determining that the first power source is unavailable (702). For
example, the one or more de-coupling components may be SCRs (e.g.,
SCRs 290, 300), and micro-controller 270 may cause SCR control unit
280 to de-assert the SCRs to de-couple the first power source from
the one or more components of the electronic device that are
connected to the output of PSM 18A (e.g., cards 12 connected via
the common Vout bus to PSM 18A). In some examples, the SCRs may be
part of a bridge rectifier 400, where the bridge rectifier is
configured to convert negative portions of an AC power of the first
power source to positive portions.
The controller may be configured to cause the ATS to de-couple the
power supply module from the first power source subsequent to
causing the one or more de-coupling components to de-couple the
first power source from the one or more components of the
electronic device (704). For example, micro-controller 270 may be
configured to cause relay driver 280 to toggle one or more relays
(e.g., relays 60, 70, 100, and 110) of ATS 16A from a close state
to an open state to cause ATS 16A to de-couple PSM 18A from the
first power source. In some example, micro-controller 270 may be
configured to wait until current through the one or more
de-coupling units is ensured to be approximately zero before
causing ATS 16A to de-couple PSM 18A from the first power
source.
The controller (e.g., micro-controller 270) may be configured to
cause ATS 16A to couple PSM 18A to the second power source (e.g.,
power source 40) (706). In this manner, one or more components 12
of electronic device 10C may receive power from power source 40
when power from power source 30 is unavailable. Micro-controller
270 and the other components of PSM 18A may implement substantially
similar techniques to transition back from power source 40 to power
source 30 in the event that power to power source 30 becomes
available again. In some examples, micro-controller 270 may be
configured to cause ATS 16A to de-couple PSM 18A from the second
power source, after a restore period, when power from the first
power source becomes available again.
In some examples, the controller may determine a ride through
period (e.g., holdup time) using the above described equation to
determine when to begin the transfer process from power source 30
to power source 40 after power source 30 is unavailable. In these
examples, the controller may cause PSM 18A to couple to power
source 40 based on the determined ride through period (e.g., the
controller may wait until the determined ride through period before
beginning the transition from power source 30 to power source 40
after power source 30 is unavailable). In some examples, the
controller may determine whether power from power source 30 is
available during the ride through period. In these examples, the
controller may de-couple power source 30 from the one or more
components of electronic device 10C if power from power source 30
is unavailable after the ride through period.
The techniques of this disclosure may be implemented in a wide
variety of devices or apparatuses, with an integrated circuit (IC)
or a set of ICs (i.e., a chip set). Various components, modules, or
units are described in this disclosure to emphasize functional
aspects of devices configured to perform the disclosed techniques,
but do not necessarily require realization by different hardware
units. Rather, various units may be combined in a hardware unit or
provided by a collection of interoperative hardware units.
Various examples have been described. These and other examples are
within the scope of the following claims.
* * * * *